Optical Flow: Optical Flow in Computer Vision

Optical flow in computer vision

Optical flow is a technique of computer vision that measures the movement of objects

Optical flow is a technique of computer vision that measures the movement of objects between successive frames of video, calculating motion vectors for each pixel. Optical flow has wide applications: from video stabilization and object tracking to motion analysis and 3D reconstruction. Optical flow can be dense (for all pixels) or sparse (for key points), depending on the application.

live demo · related simulation● LIVE

FAQ: Questions and Answers

Frequently asked questions

How is optical flow calculated?

Optical flow calculates motion vectors for pixels between frames, using classical methods (Lucas-Kanade, Horn-Schunck) or deep learning models (FlowNet, RAFT). It shows how objects move in video.

What does optical flow do?

Optical flow algorithms determine the direction and speed of movement within a visual scene by analyzing changes between consecutive frames of an image or video.

This section provides information about optical flow, a core technique in computer vision for understanding motion within visual data.

▶ Try it live

Everything above runs in your browser — open Gradient Descent Visualiser and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.

▶ Open Gradient Descent Visualiser simulation

Optical Flow: Optical Flow in Computer Vision | AI Knowledge Hub

Optical flow in computer vision

Optical flow is a technique of computer vision that measures the movement of objects

FAQ: Questions and Answers

Frequently asked questions

▶ Try it live

What did you find?